Optimizing Screening for Tuberculosis and Hepatitis B Prior to Starting Tumor Necrosis Factor-alpha Inhibitors in Crohn's Disease

Background and Aims Treatment with tumor necrosis factor-alpha (TNF-alpha) inhibitors in patients with Crohn's disease (CD) is associated with potentially serious infections, including tuberculosis (TB) and hepatitis B virus (HBV). We assessed the cost-effectiveness of extensive TB screening and HBV screening prior to initiating TNF-alpha inhibitors in CD. We constructed two Markov models: (1) comparing tuberculin skin test (TST) combined with chest X-ray (conventional TB screening) versus TST and chest X-ray followed by the interferon-gamma release assay (extensive TB screening) in diagnosing TB; and (2) HBV screening versus no HBV screening. Our base-case included an adult CD patient starting with infliximab treatment. Input parameters were extracted from the literature. Direct medical costs were assessed and discounted following a third-party payer perspective. The main outcome was the incremental cost-effectiveness ratio (ICER). Sensitivity and Monte Carlo analyses were performed over wide ranges of probability and cost estimates. At base-case, the ICERs of extensive screening and HBV screening were a,not sign64,340 and a,not sign75,760 respectively to gain one quality-adjusted life year. Sensitivity analyses concluded that extensive TB screening was a cost-effective strategy if the latent TB prevalence is more than 12 % or if the false positivity rate of TST is more than 20 %. HBV screening became cost-effective if HBV reactivation or HBV-related mortality is higher than 37 and 62 %, respectively. Extensive TB screening and HBV screening are not cost-effective compared with conventional TB screening and no HBV screening, respectively. However, when targeted at high-risk patient groups, these screening strategies are likely to become cost-effective